Ebook Basic transesophageal and critical care ultrasound: Part 2

PART II


Chapter 13
Critical Care Ultrasound Examination of the
Nervous System
Andrea Rigamonti, Robert Chen, Ramamani Mariappan and
Céline Odier

INTRODUCTION
Monitoring cardiovascular and respiratory function is an important aspect of patient
management in the intensive care unit (ICU) and operating room. Cardiovascular and
pulmonary functions are continuously monitored during anesthesia and in critical care,
but the monitoring of cerebrovascular function is not routinely performed. With the
availability of non-invasive monitors like transcranial Doppler (TCD), quantitative
electroencephalographic monitoring and near-infrared spectroscopy, continuous
monitoring of cerebrovascular function is possible. Patient care can be enhanced when
the information collected from these monitors are used to guide patient management. In
this chapter, the role of ultrasound (US) in examining the central nervous system,
including TCD, US of the optic nerve, and direct visualization of the brain using 2D
echography will be presented.

TRANSCRANIAL DOPPLER ULTRASOUND
Transcranial Doppler US is a simple, non-invasive, relatively cheap bedside tool that
can provide real-time dynamic information regarding cerebral blood flow velocity in
the basal cerebral blood vessels. Since the first clinical application in 1982, 1 the use of
TCD has expanded rapidly over the past two decades. The portability and non-invasive
nature of TCD allows both monitoring during emergencies and serial monitoring in the
ICU. The clinical applications of TCD are summarized in Table 13.1. Transcranial
Doppler is currently used in neuro-critical care units, acute stroke units, operating
rooms, emergency departments, and even in outpatient settings to assess the

hemodynamic changes associated with stenosis of large cerebral arteries or to


determine patients at risk of stroke with sickle cell disease. For the experienced
vascular neurologist, neuro-intensivist, and neuro-anesthesiologist, the small portable
TCD device serves as a “stethoscope for the brain”. 2
Table 13.1 Applications of Transcranial Doppler

BASIC PRINCIPLES OF TRANSCRANIAL DOPPLER
The TCD probe works only using Doppler signals and does not acquire 2D imaging.
It emits a range gated, pulsed-wave Doppler US beam at a low (2 MHz) frequency.
The US beam penetrates the skull at areas called “acoustic windows” and is scattered in
the tissue. Some of the US wave are reflected back at an altered frequency by the
moving red blood cells. The difference in frequency between the transmitted and
received sound waves is called the “Doppler frequency shift” (Fd) or “Doppler effect”.
The reflected waves are received by the Doppler probe and transformed into an
electrical signal. The computer performs a fast Fourier analysis to transform this
electric signal into a moving graphic display with the time on the x-axis and the blood
flow velocity on the y-axis (see Chapter 2, Patient Safety and Imaging Artifacts). Apart
from insonation angle, other factors such as the vessel diameter, hematocrit, arterial
carbon dioxide tension (PaCO2), blood pressure, body temperature, and the presence of
collateral flow can also affect the cerebral blood flow velocity (CBFV). Some
epidemiologic and physiologic factors such as age, gender, pregnancy, and sleep-awake


pattern can also affect the CBFV. These should all be kept in mind while interpreting the
CBFV in various clinical situations. 3

Fig. 13.1 Transcranial Doppler devices. Specialized transcranial Doppler monitoring devices are shown: (A) ST3
(Spencer Technology, Seattle WA) and (B) Sonara (Natus Medical, San Carlos, CA, USA)

Fig. 13.2 Power motion (M)-mode Doppler. Diagram shows interrogation of cerebral vessels with power M-mode or
combined color Doppler and M-mode transcranial Doppler (TCD). The ultrasound probe is positioned over the left
temporal region. The TCD display shows an upper portion in red, which corresponds to flow in the ipsilateral left
middle cerebral artery (LMCA). The middle blue portion is associated with the ipsilateral left anterior cerebral artery
(LACA) Doppler signal moving away from the transducer. The lower red portion corresponds to flow in the
contralateral right anterior cerebral artery (RACA)

DEVELOPMENTS IN TRANSCRANIAL DOPPLER
TECHNOLOGY
There are several measurements in TCD using either specialized equipment (Figure
13.1 ) or the basic transthoracic probe. These modalities include:
Continuous and pulsed wave technique, which are described in Chapter 1,
Ultrasound Imaging: Acquisition and Optimization.
Power motion-mode Doppler (PMD/TCD): Moehring and Spencer 4 introduced
this mode of Doppler technique in 2002. This modality displays all available


flow signals and direction over a range of 6 cm of intracranial space
simultaneously in a single spectral display (Figure 13.2). Time spent for TCD
examination is reduced compared to a single channel spectral TCD. This mode
simplifies the TCD examination for the inexperienced operator.
Transcranial color-coded duplex sonography (TCCS): This mode combines
pulsed-wave Doppler with two-dimensional, real-time B-mode imaging
(Figure 13.3). Transcranial color-coded duplex sonography allows the
visualization of all basal cerebral arteries through the intact skull and allows
precise placement of the Doppler sample volume in the vessel. Transcranial
color-coded duplex sonography is more reliable and accurate in the detection of
pathological hemodynamic changes than conventional TCD for intracranial
arteries, other than middle cerebral artery (MCA) or in the setting of anatomical

distortions from tumor, hematoma, and edema displacing normal structures. 4 – 6

Fig. 13.3 Transcranial Doppler color-coded duplex sonography (TCCS). The middle cerebral artery (MCA) is
interrogated using a transthoracic probe positioned over the right temporal region. (A) A 2D image of cerebral artery
structure and color Doppler (Nyquist 36 cm/s) flow interrogation is obtained. (B) Sample volume positioning in the
vessel allows precise determination of the MCA velocity spectral Doppler profile. (C) In this patient, half of the Circle
of Willis is imaged using TCCS. (D) Spectral Doppler profile of the anterior cerebral artery shows the velocity
direction is away from the transducer. HR, heart rate


A: />
B: />
ACOUSTIC WINDOWS
In order to interrogate the brain, it is essential to obtain an acoustic window through the
skull. Normally, the US waves undergo gradual loss of intensity as they move through
different body structures. The degree of attenuation is directly proportional to the
attenuation coefficient of the medium and to the US frequency. Since bone has a
relatively high attenuation coefficient, it is difficult to measure CBFV using a
conventional 5–10 MHz Doppler probe. The use of a lower frequency (1–2 MHz) probe
is required. Transcranial Doppler examinations are commonly performed through four
“acoustic windows” where the bone is relatively thin or absent. The orientation of the
ultrasound probe in each acoustic window is shown in Figure 13.4. The depth,
direction of blood flow, and the CBFV of the vessels insonated in each window are
shown in Table 13.2. Table 13.3 summarizes how to perform TCD .


Fig. 13.4 Acoustic windows and ultrasound probe position. Lateral skull diagram showing probe positions used to
obtain acoustic windows for transcranial Doppler: (1) trans-orbital, (2) submandibular, (3) suboccipital or
transforaminal, and (4) transtemporal. (Anatomical images with permission of Primal Pictures, Wolters Kluwer
Health.)

Table 13.2 Normal Doppler Values


ACA, anterior cerebral artery; C, carotid segments (C1, cervical or
submandibular segment; C2,petrous segment; C3, lacerum segment; A1, ACA
first horizontal segment; C4, cavernous segment; C5, clinoid segment; C6,
ophthalmic segment; C7, communicating or terminal (t) segment); ED, enddiastolic; ICA, internal carotid artery; MCA, middle cerebral artery; P1, PCA
first horizontal segment; P2, PCA second horizontal segment; PCA, posterior
cerebral artery; PI, pulsatility index; RI, resistance index; TICA, terminal
internal carotid artery or C7. (Adapted from Rigamonti et al. 7 )
Table 13.3 General Procedural Steps in Echo-Guided Transcranial Doppler


ACA, anterior cerebral artery; ACoA, anterior communication artery; C, carotid
segments (C1, cervical or submandibular segment; C2, petrous segment; C3,
lacerum segment; C4, cavernous segment; C5, clinoid segment; C6,
ophthalmic segment; C7, communicating or terminal (t) segment); MCA,
middle cerebral artery; MI, mechanical index; P1, PCA first horizontal
segment; P2, PCA second horizontal segment; PCA, posterior cerebral artery;
PCoA, posterior communicating artery; SPTA, spatial peak temporal average;
TICA, terminal internal carotid artery.Trans-temporal window: The probe is
placed over an area just above the zygomatic arch represented by a line
joining the tragus to the lateral canthus of the eye. There are four locations
within the trans-temporal window: anterior, middle, posterior, and frontal
(Figure 5). The MCA, anterior cerebral artery (ACA), posterior cerebral artery
(PCA), and internal carotid artery (ICA) can be interrogated (Figure 6).
Reference points using 2D imaging, are the petrous bone, foramen lacerum,
sphenoid wing, and the opposite cranial wall (Figure 7). In order to see the
latter, the depth has to be adjusted to at least twice the distance from the

midline cerebral falx which is typically at 8 cm.
1. Trans-temporal window: The probe is placed over an area just above the
zygomatic arch represented by a line joining the tragus to the lateral canthus of
the eye. There are four locations within the trans-temporal window: anterior,
middle, posterior, and frontal (Figure 13.5). The MCA, anterior cerebral


artery (ACA), posterior cerebral artery (PCA), and internal carotid artery
(ICA) can be interrogated (Figure 13.6). Reference points using 2D imaging,
are the petrous bone, foramen lacerum, sphenoid wing, and the opposite
cranial wall (Figure 13.7). In order to see the latter, the depth has to be
adjusted to at least twice the distance from the midline cerebral falx which is
typically at 8 cm.
2. Transorbital window: The probe is placed over the upper eyelid to insonate
the ophthalmic artery (OA) and portions of ICA (cavernous, genu, and
supraclinoid), across the carotid siphon. While measuring the CBFV through
this window, the ultrasound power has to bedecreased to the minimum (10%)
to avoid thermal injury to the retina (Figure 13.8).
3. Suboccipital or transforaminal window: In this window the terminal portion of
the vertebral arteries (VA) and the basilar artery (BA) are insonated. The
probe is initially placed in the midline over the upper part of posterior neck
(2.5 cm below the skull edge), while the patient is sitting or lying in the lateral
position. This approach facilitates insonation of the BA, while moving the
probe 2.5 cm lateral from the midline on each side identifies the VA ( Figure
13.9).
4. Sub-mandibular window: The probe lies below the angle of the mandible to
insonate the extra-cranial portion of ICA. Anatomic features of the patient may
make the differentiation of the ICA from the external carotid artery (ECA)
challenging. However, typically the diastolic component of the ICA is more
apparent than the ECA because of increased resistance of the

muscularterritories irrigated by the ECA (Figure 13.10).


Fig. 13.5 Temporal windows. The four locations for probe position within the transtemporal window for the (A)
middle cerebral artery (MCA); (B) bifurcation of the MCA and anterior cerebral artery (ACA); (C) ACA; (D)
terminal internal carotid artery (TICA); (E) pre-communicating posterior cerebral artery (PCA); and (F) postcommunicating PCA. (Anatomical images with permission of Primal Pictures, Wolters Kluwer Health.)

Fig. 13.6 Transcranial Doppler signals. Probe position in the temporal window and normal transcranial Doppler signals
are shown for the (A) middle cerebral artery (MCA); (B) bifurcation of the MCA and anterior cerebral artery (ACA);
(C) ACA; (D) terminal internal carotid artery (TICA); (E) pre-communicating posterior cerebral artery (PCA); and
(F) post-communicating PCA. (Anatomical images with permission of Primal Pictures, Wolters Kluwer Health.)


Fig. 13.7 Temporal windows. (A,B) Using 2D imaging, anatomic reference points shown with these cut portions of
the skull are the petrous bone, foramen lacerum, sphenoid wing, and the opposite cranial wall (arrows). (C) Color
Doppler (Nyquist 27 cm/s) showing blood flow in the petrous bone (arrows). The sphenoid wing is shown (triangles).
(D) The display depth is initially adjusted in order to see the opposite skull (arrows). (Anatomical images with
permission of Primal Pictures, Wolters Kluwer Health.)

D: />

Fig. 13.8 Orbital window. Probe positions in the orbital window and transcranial Doppler signals are shown for
different segments of the internal carotid artery (ICA). The most distal portion is the (A) ophthalmic artery that
originates from the ophthalmic segment. From distal to proximal the segments of the ICA are the (B) supraclinoid
segment, (C) petrous segment that includes the bend or genu, and (D) parasellar segment or cavernous carotid siphon.
(Anatomical images with permission of Primal Pictures, Wolters Kluwer Health.)

B: />
Transcranial Doppler Indices
The peak systolic flow velocity s(FVs) and the end-diastolic flow velocity (FVd) are

measured directly by analyzing the waveform. The mean flow velocity (FVm),
pulsatility index (PI) and resistance index (RI) are estimated from these measured
values using the following formulas. Ultrasound machines with automatic or manual
spectral waveform tracing calculate FVm as the area under the traced curve.


Fig. 13.9 Occipital window. Probe positions in the occipital window and transcranial Doppler signals are shown for
the (A) basilar artery, (B) anterior inferior cerebralartery, and (C) posterior cerebral artery. (Anatomical images with
permission of Primal Pictures, Wolters Kluwer Health.)

/>
Fig. 13.10 Submandibular window. (A) 2D image with color Doppler (Nyquist 9 cm/s) shows systolic flow in both the


external carotid artery (ECA) and internal carotid artery (ICA). (B) 2D image with color Doppler (Nyquist 9 cm/s)
shows more prominent diastolic flow in the ICA, helping to distinguish the ICA and ECA. (C,D) Spectral Doppler
trace confirms systolic flow in the ECA and continuous flow with systolic predominance in the ICA

A&B: />
C&D: />
Mean flow velocity FVm = FVs + 2 x FVd /3 or = FVs - FVd /3 + FVd
Pulsatility index (PI) = (FVs - FVd)/FVm
Resistance index (RI) = (FVs - FVd)/FVs
Table 13.2 shows the normal mean velocities, PI, RI, and the diameter of basal
cerebral vessels. Arterial velocity normally follows this order of magnitude: MCA >
ACA ≥ Siphon ≥ PCA ≥ BA > VA > OA. A simple rule for mean velocity (in cm/s) is
MCA 60, ACA, and TICA 50, PCA 40, BA 30, and VA 20. The The difference among
the MCA, ACA, and PCA velocities should velocities should be <30%. In a normal
MCA, the FVm should not exceed 170 cm/s in children and 80 cm/s in adults. With
normal breathing, the normal end-diastolic velocity should be 25–50% of the peak

systolic velocities and the PI should be low (0.6–1.1), except for the OA (PI >1.2).

Limitations of Transcranial Doppler
1. Transcranial Doppler is highly operator dependent and requires detailed 3D
knowledge of cerebrovascular anatomy and good knowledge of confounding
factors affecting CBFV.
2. Transcranial Doppler is impossible or very difficult in 8–10% of patients
because of inadequate temporal windows. Poor to absent temporal window is
more common in those of African descent, Asians, and elderly female patients.
This is related to thickness and porosity of the temporal bone attenuating
ultrasound energy transmission.
3. The direction of blood flow can vary; altering the interpretation of TCD. It has
been found that in more than 50% of healthy brains and in more than 80% of
dysfunctional brains, the Circle of Willis contains at least one artery that is
absent or underdeveloped. 8 , 9 Anatomical variants are described in up to


52%. 10

Applications of Transcranial Doppler
Aneurysmal Subarachnoid Hemorrhage
Cerebral vasospasm leading to delayed cerebral ischemia (DCI) and infarction is the
most common and devastating complication following an aneurysmal subarachnoid
hemorrhage (aSAH). Vasospasm can affect either the stem of major intracerebral
vessels, distal vessels, or both. Cerebral vasospasm usually peaks at 3–7 days
following aSAH and can last for 10–14 days. Following aSAH, 14–46% of patients
develop DCI, of which 64% will develop infarction due to severe narrowing (<1 mm)
of intracranial arteries. 2 For the past two decades, there has been a significant
reduction in mortality following aSAH, due to improved vasospasm surveillance
technologies. Digital subtraction angiography, computed tomography angiogram (CTA),

and computed tomography perfusion (CTP) are very useful tools for the diagnosis of
vasospasm. These techniques give snap-shot pictures rather than a continuous
assessment and expose the patient to radiation and contrast. Transcranial Doppler is a
simple, non-invasive, continuous, bedside tool for monitoring vasospasm without
exposing the patient to radiation or contrast. According to the 2008 guidelines of the
American Academy of Neurology, TCD is accepted as a tool for diagnosing and
monitoring cerebral vasospasm (Recommendation A/I to II). A recent meta-analysis, 11
comparing TCD with angiography showed that TCD reliably predicted MCA
vasospasm in 97% of SAH with a specificity of 99% and a sensitivity of 67%.
However, there is poor sensitivity and specificity in evaluating vasospasm in the ACA
and PCA territory. As mentioned previously, several factors such as mean arterial
pressure (MAP), PaCO2, hematocrit, collateral flow pattern, response to therapeutic
intervention, intracranial pressure, age, and technical error can all affect the TCD
velocities. While interpreting the results, all those factors have to be considered before
making a definitive conclusion.
The mean blood flow velocity increases when vasospasm involves a proximal vessel
(Figure 13.11). Whereas when vasospasm involves the distal part of the intracranial
arteries, the mean blood flow velocity does not increase, but blood flow creates focal
pulsatility, that increases the PI to >1.2. Studies have shown that an increase in MCA
mean velocity by >25 to 40 cm/s per day or by 50% from the baseline in 24 hours is an
indicator of vasospasm. In order to follow these criteria, the baseline MCA velocity at
admission should be measured for comparison. MCA vasospasm can be graded as mild,
moderate, and severe according to the MCA blood flow velocity (Table 13.3). Studies
have shown a correlation between TCD grading and angiographic vasospasm.
The Lindegaard Index (LI) are a set of given criteria used to differentiate the increase
in flow velocity caused by hyperemia from vasospasm by comparing intra- to extra-


cranial blood flow velocities (Figure 13.12).
LI = FVm in MCA/FVm in extracranial portion of ipsilateral ICA.

When increased flow velocity is due to hyperemia, it affects both intracranial arteries
and the extracranial portion of the ICA, therefore the ratio will be low (typically <3).
When vasospasm is the cause of high flow velocity, the extracranial portion of the ICA
is unaffected; therefore the Lindegaard index will be elevated. Transcranial Doppler is
very specific for for diagnosing vasospasm in the posterior circulation. 12 A modified LI
is used to differentiate hyperemia from vasospasm in the posterior circulation with a
value >2 indicating vasospasm. 13
Modified LI = FVm (BA)/FVm (VA)
Several authors have combined different TCD indices to increase the sensitivity and
specificity. Gonzalez et al. 14 combined the TCD velocity and with ipsilateral
hemispheric blood flow using Xenon and created a vasospasm index, where a value of
>3.5 indicates vasospasm.
Vasospasm Index = TCD velocity/hemispheric blood flow Nakae et al. 15 combined
the ipsilateral and contralateral mean blood flow velocity using TCD. A ratio between
both of >1.5 predicted delayed cerebral ischemia more accurately than absolute blood
flow velocity alone. Finally, if there is a carotid artery stenosis, velocities will also be
increased proportional to the degree of the stenosis (Table 13.4).

Fig. 13.11 Vasospasm. Increased right middle cerebral artery (R-MCA) velocities are shown with transcranial
Doppler in a patient with cerebral vasospasm following aneurysmal subarachnoid hemorrhage. Both normal peak (80–
120 cm/s) and mean (62 ± 12 cm/s) velocities are exceeded. EDV, end-diastolic velocity; HR, heart rate; PI, pulsatility
index


A: />
C: />
Fig. 13.12 Vasospasm algorithm. An algorithm for the differentiation of cerebral vasospasm versus hyperdynamic
flow using mean middle cerebral artery (MCA) velocities and the Lindegaard Index is presented. For basilar artery
vasospasm, the criteria are: basilar artery velocity > 85 cm/s + LI (basilar artery/MCA) > 2-3

Brain Death
Brain death is defined as “an irreversible loss of brain function including the
brainstem”. The American Academy of Neurology has published diagnostic
requirements for confirming brain death by clinical criteria. In certain clinical situations
where brain death determination cannot be reliably performed by clinical criteria,
confirmatory tests are mandatory. These situations include severe facial trauma, preexisting pupillary abnormalities, toxic levels of sedative drugs, aminoglycosides,
tricyclic antidepressants, anticholinergics, antiepileptics, chemotherapeutic agents or
neuromuscular blocking agents, metabolic disturbances, hypothermia, conditions like
severe sleep apnea or severe cardiorespiratory disease, or the inability to correctly
perform an apnea test. Absence of cerebral blood flow demonstrated by four-vessel
digital angiography is considered the gold standard for diagnosing brain death. Other
ancillary radiological tests, such as CTA, CTP and magnetic resonance angiography are
also used to confirm brain death. These tests are invasive, expensive, require transport


of critically ill patients, and, for iodinated contrast studies, have additional contrastinduced complications.
Transcranial Doppler has been used for the diagnosis of cerebral circulatory arrest
(CCA) since 1987 16 , 17 with a high sensitivity (91–100%) and specificity (100%).
There have been several reports of positive CBF on TCD, in patients who are clinically
brain dead. In these cases, the brain insult or damage was limited to the cerebellum or
brainstem leaving blood flow in the anterior cerebral circulation relatively intact. For
this reason, it is important to obtain bilateral MCA and BA flow patterns on two
occasions at an interval of 30 minutes before diagnosing CCA by TCD. Infants (open
fontanelle) and patients who underwent decompressive craniotomy with an
intraventricular drainage catheter, will still have detectable CBF even in the presence
of clinical brain death (false negative). Following transient hypotension and cardiac
arrest, TCD findings can be similar to CCA findings (false positive). Caution is needed
before interpreting the TCD results in this situation.
Table 13.4 Doppler Spectral Criteria to Evaluate Carotid Stenosis

In the presence of ICA occlusion: unilateral dampened flow will be observed
in the CCA. In addition, absent or reversed diastolic flow proximal to ICA
occlusion will be present. CCA, common carotid artery; ICA, internal carotid
artery. (Adapted from Carroll et al. 12 )


Fig. 13.13 Intracranial hypertension and circulatory arrest. Transcranial Doppler changes in middle cerebral artery
(MCA) mean flow with progressive increase in intracranial pressure (ICP) are shown compared with (A) normal
MCA flow trace and normal ICP. (B, C) The initial stage has a typical pattern of systolic peaks with progressive
reduction in diastolic velocities. (D—G) The three patterns that correspond to intracranial circulatory arrest are shown:
biphasic oscillating flow, systolic spike flow, and zero flow. DAP, diastolic arterial pressure; SAP, systolic arterial
pressure. (Adapted from Hassler et al. 18 and Conti et al. 19 )

Flow Patterns with Increased Intracranial Pressure Leading to
Cerebral Circulatory Arrest
Three types of flow patterns are noted in the Doppler spectral wave form in cerebral
circulatory arrest (Figure 13.13). First, the oscillating or reverberant flow pattern
represents systolic flow towards the brain and a diastolic flow away from the brain so-called ‘bidirectional’ or ‘reverberant’ flow ( Figure 13.13 D,E). Oscillating flow is
defined by signals with forward and reverse flow component in one cardiac cycle. The
area under the curve of both the antegrade and retrograde flow pattern (to-and-fro
movement) should be identical. In this situation, extensive ischemia, intracranial
bleeding, or brain swelling can severely increase intracranial pressure (ICP). When ICP
reaches the level at which it obstructs the microcirculation, forward flow during systole
expands the arterial tree, but due to the very high distal resistance, little or no flow
occurs through the microcirculation. The second pattern is the systolic spike pattern that
occurs when the ICP reaches the diastolic pressure (Figure 13.13 C). The peak intensity
of the systolic spike should be <50 cm/s and the duration <200 ms without a flow signal
during the remaining cardiac cycle. Finally, the third pattern corresponds to the “no
flow pattern” (Figure 13.13 F,G). When ICP reaches MAP, there will be no flow in the
major intracranial arteries. Since there is an absent acoustic window in 10% of patients,

there should have been an initial documented flow pattern before interpreting this
pattern of CCA. This can be associated with disappearance of diastolic flow in the
extra-cranial segment of the ICA and tendency to evolve toward the oscillating flow.
Note that the TCD changes observed with a progressive increase in ICP are almost
identical to those described with circulatory arrest. The aspect of the TCD signals will
also be influenced by the underlying cardiac pathologies and devices such as intraaortic balloon pump (Figure 13.14).

Hyper-Intensity Thromboembolic Signals
Cerebral embolism is common during carotid endarterectomy (CEA), coronary artery
bypass graft (CABG) surgery, cardiac valve surgeries, and aortic surgery. This
embolism produces characteristic hyper-intensity thromboembolic signal (HITS) on
TCD examination. The duration and relative increase in intensity of the HITS from the
background signal correlates with the size of emboli (Figure 13.15).
For demonstration of HITS, all of the following criteria are required: (1) random


occurrence; (2) brief duration (0.01–0.1 s); (3) high intensity (3 dB above the
background intensity); (4) primarily unidirectional quality within the Doppler spectrum;
(5) causing a spike in the power/intensity trace; and (6) accompanied by an audible
“chirp” or “pop.” Hyper-intensity thromboembolic signals can be easily differentiated
from artifacts created by probe movement, patient movement, and electrical interference
by its low frequency, non-harmonic quality. The high-risk periods for embolic stroke in
patients undergoing CEA include shunt insertion, carotid cross-clamp application and
release, wound closure, and during the initial 12-hour, post-operative period. A finding
of >10 HITS during any phase of surgery, >5 during any 15-minute period in the
recovery room, or >50 HITS per hour during the postoperative phase is predictive for
the development of cerebral infarction following CEA.
Postoperative neurological complications significantly alters recovery after cardiac
surgery. 20 The incidence of clinically apparent periprocedural strokes are estimated to
occur in 1.6–6.1% of patients undergoing cardiac surgery. 21 – 23 Several factors

increase this risk, including the presence of extracranial ICA stenosis, a history of
previous stroke, and a prolonged bypass pump time. Hyper-intensity thromboembolic
signal are common at the aortic cross-clamp application and removal during CABG.
The number of HITS is even higher during cardiac valve surgeries. Hyper-intensity
thromboembolic signal occurring in patients with prosthetic valves often have
intensities exceeding 24 dB with durations >50 ms. With the availability of ambulatory
TCD (similar to the Holter monitoring system), continuous TCD monitoring is possible
for up to 8 hours. This can assess the true embolic load and predict the stroke risk,
especially for asymptomatic carotid stenosis of >50%.

Fig. 13.14 Transcranial Doppler (TCD) display. (A) Anterior cerebral artery velocity in an elderly patient with severe
aortic stenosis. Note the delayed upstroke or pulsus tardus (dotted line) that was bilateral. When unilateral, carotid
stenosis should be suspected. (B) Right middle cerebral artery TCD signal with an intra-aortic balloon pump (IABP)
turned on and off are shown


Fig. 13.15 Hyper-intensity thromboembolic signal (HITS). Note the significant number of HITS in both the M-mode
and Doppler signals (arrows) demonstrated with TCD during a percutaneous aortic valve procedure. The HITS
appeared during guidewire positioning across the ascending aorta. The total number of HITS was 719 on the right
middle cerebral artery (RMCA) and 922 on the left middle cerebral artery (LMCA)

Transcranial Doppler can help assess the risk of stroke in asymptomatic carotid
stenosis, 24 the adequacy of anticoagulation in patients with acute embolic stroke in the
proximal cerebral arteries and in patients with prosthetic cardiac valves. There are,
however, several limitations in using TCD as a monitor of embolic stroke. First, it is
very difficult to differentiate between embolic materials containing air, atheromatous
plaques, lipid or platelet aggregates. In order to distinguish gaseous emboli from solid
emboli during carotid surgery, inspiring 100% oxygen reduces the HITS rate by >90%,
indicating that most of the embolic signals are from gaseous origin. These embolic
signals produce a low frequency sound and have high reflectivity, the signal goes

beyond the waveform of the Doppler spectrum, while solid emboli signals are
contained within the waveform of Doppler spectrum.

RAISED INTRACRANIAL PRESSURE IN PATIENTS
WITH SEVERE HEAD INJURY
Cerebral perfusion pressure (CPP)-guided management with ICP monitoring is the
standard of care in patients with severe head injury. Intraventricular ICP monitoring is
considered the gold standard for measuring ICP, although with the potential risks of
infection, hemorrhage, malposition, and malfunction. In certain clinical situations, the


use of invasive ICP monitoring (both intraventricular and intraparenchymal) is not
feasible either because of non-availability of personnel and/ or instruments or it is too
risky to perform (coagulopathy or very small or compressed ventricles). In order to
circumvent these limitations, two alternative techniques can be used, optic nerve sheath
diameter (ONSD) and TCD monitoring.

Optic Nerve Sheath Diameter
It has been said by Biblical scholars and poets that “the eye is the window of the soul.”
Certainly, with the advent of fundoscopy, the eye became a window to the brain. The
association between papilledema and raised ICP is well known (Figure 13.16). As
cerebrospinal fluid (CSF) pressure increases, so does the pressure in the optic nerve
and its sheath. The resultant edema is the bulging noted as the nerve inserts into the
retina. Logically, the optic nerve diameter may change with the edema caused by
increased CSF pressure. The use of portable ocular ultrasound has the potential value of
detecting increased ICP by measuring the ONSD. The optic nerve sheath is a
continuation of dura mater around the optic nerve. The perineural space of the
intraorbital portion of the optic nerve, which is the space between the optic nerve sheath
and optic nerve, is in direct communication with the subarachnoid space of the brain.
This portion of the optic nerve is directly affected by changes in ICP When ICP rises

above 20 mmHg, CSF is displaced into the perineural space of the optic nerve,
increasing the nerve sheath diameter. Several clinical studies have proven that
millimetric increase in ONSD directly correlates with increasing ICP values. The
receiver operating characteristics of an ONSD >5 mm as a cut-off to detect an ICP
above 20 cm H2O is 0.93. 25

Fig. 13.16 Papilledema. (A,B) Ultrasound of the eye orbit showing 2D images of papilledema (arrow) in two brain
dead patients for organ donation

A: />

Fig. 13.17 Optic nerve examination. (A) Photo of high frequency ultrasound probes that can be (B) gently positioned
over the eyelid of a closed eye. (C) A 2D image easily displays the ocular structures, including the lens, posterior
chamber, and optic nerve sheath

C: />
Fig. 13.18 Optic nerve sheath measurement. The site for measuring the diameter of the optic nerve sheath is shown.
A 3 mm perpendicular line is drawn from the middle of the optic nerve, at which point the transverse measurement of
the optic sheath is performed. Note that the measurement includes the sheath and stops at the transition contrast
between the optic nerve and surrounding tissue. Inset shows the optic nerve sheath measurement in a comatose
patient

Measurement Technique for Optic Nerve Sheath Diameter
The examination is performed through the eyelid, which protects the ocular globe from
abrasion. The liquid-filled globe is an excellent conductor of US much as a full bladder
is for a pelvic US. A 7.5-10 MHz probe is placed without pressure over the closed
upper eyelid after applying an adequate amount of US gel. The transmitted US power is
reduced to 50%. The Bromage grip (Figure 13.17) is used; the examiner’s hand or wrist

rests on the patient’s cheek or forehead supporting the weight of the probe allowing for
more delicate probe manipulation. The frequency chosen should be the highest that will
allow visualization of the optic nerve and sheath. The nerve and nerve sheath have a
distinctive US texture that is different from the rest of the posterior globe. It is imaged
as a hypoechoic structure extending from the retina posteriorly. The optic nerve sheath
is subtly more echogenic and surrounds the nerve (Figure 13.17). As with all US
examinations, the region of interest is placed in the middle of the US display with the
focus. Given the average globe size of 2.5 cm, an ONSD examination rarely requires
more than 4–5 cm of total depth. Authors have described scanning the ONSD in both an
axial and saggital plane. 26 The differences measured are in the submillimeter range. Our
local standard is to scan only in the axial plane.
The ONSD is measured 3 mm deep to the retina (Figure 13.18). Measurements
should be taken in each eye and averaged to obtain the binocular ONSD. In case of
unilateral eye injury, only one eye measurement can be taken. We advocate saving a still
image and measuring off-line. Such a practice prevents distracted scanning of the orbit
and possible application of undue pressure as well as reducing the difficulty of
measuring a moving target. If available, a “sweep” through the orbit could be saved as a
cine loop on the US machine allowing the user to freeze the image where the largest
ONSD is noted. Care must be taken to avoid the measurement of artifacts. 27 Studies
have shown that an ONSD measurement of >5 mm is considered abnormal with a
sensitivity of 100% and specificity of 65%. In pediatric, head-injured patients, a
reported cut-off value of >4.5 mm in children older than 1 year was considered to be
abnormal. The use of color Doppler to identify retinal vessels can improve the
accuracy. 28 In addition, retinal vessel velocity has been shown to correlate with
systolic blood pressure and can be used to determine the presence of flow and confirm
the patency of the Circle of Willis (Figure 13.19). 29 The major advantages of OSND
technique are its simplicity, portability, non-invasiveness, and low cost, which allow
repeated measurements without the risk of transportation. It is important to note that the
CT findings will be normal during the early stages of head injury. Serial US
examination can reveal ICP changes in these patients and can guide further management

before the secondary insult occurs. Combining both OSND and TCD gives more insight
into brain pathophysiology (Figure 13.20).